First edits of literature review

[turgeonmaxime]

I've reviewed the literature review from @Jesse-Islam . Before we can merge, there are a few things we need to address. Some of them appear as comments in red in the text.

• [x] We need to explain how we went from 60 packages identified by the scraper to 14 packages in the table and the text, because that's a big drop.
• [ ] Is it true that casebase is the only package doing penalized parametric modelling? This would be surprising to me.
• [ ] There seems to be some packages missing from the discussion on competing risks. It may help to look at the vignette from the CFC package.
• [ ] The package names are not consistent throughout. For example, the survival package is also sometimes called Survival. This should be fixed by always using the official name (e.g. survival), both in the text and in the table.

@Jesse-Islam : I'll take care of the third point on competing risks, but could you address the other three points? You can use git checkout review to make the changes on the right branch. I'll continue with my other edits on a new branch.

[Jesse-Islam]

1. the 60 to 14 packages were scanned by Sahir and myself and was not a systematic process... I could make it systematic by scrubbing through their reference manuals (PDFS) and building up a table for all 60, where as the ones discussed would be the ones we found particularly interesting. Would that be better for the paper?

2. from the 14 chosen, rstpm2 also does penalized parametric modelling. It seems my writing does not particularly highlight this, so I'll fix that! its possible others do as well but our 60 to 14 filtering may not have caught it (since not everything the package does is necessarily in the CRAN page description).

3. I can manually code a few things to reduce the number of packages I use.

4. Ah yes I noticed that, I wrote it on my notes but forgot to add it to the commit summary. I will take care of this!

[turgeonmaxime]

Excellent, thanks for the clarification! For point 1, it's fine if it wasn't too systematic, but I think we should add a few details about what you guys were looking for during the manual curation. E.g. you looked at the vignettes, or the functions, to see which of the 60 packages were false positives for the characteristics we were screening.

For point 3, I'll take care of this, don't worry about competing risks.

[turgeonmaxime]

I'll merge this pull request. I think the case studies are in good shape. I'll work on the discussion and the intro on a new branch, so you can guys can have a look at those case studies if you want.